node is running but the repmgr node record is inactive

162 views
Skip to first unread message

dcvyth...@gmail.com

unread,
May 28, 2021, 7:13:01 AMMay 28
to repmgr
Hello,

I have a two node replication cluster, it's working fine for about 10 months now. I have switched-over many times with no issues.
Today while looking for other things, I noticed both nodes respond that:


postgres@cn1:~$ repmgr -f /etc/repmgr.conf cluster show
 ID | Name            | Role    | Status    | Upstream | Location | Priority | Timeline | Connection string                                                  
----+-----------------+---------+-----------+----------+----------+----------+----------+-----------------------------------------------
 1  | cn1.example.com | primary | * running |          | default  | 100      | 16       | host=cn1.example.com user=repmgr dbname=repmgr
 2  | cn2.example.com | standby | ! running |          | default  | 100      | 16       | host=cn2.example.com user=repmgr dbname=repmgr

WARNING: following issues were detected
  - node "cn2.example.com" (ID: 2) is running but the repmgr node record is inactive

postgres@cn2:~$ repmgr -f /etc/repmgr.conf cluster show
 ID | Name            | Role    | Status    | Upstream | Location | Priority | Timeline | Connection string                                                  
----+-----------------+---------+-----------+----------+----------+----------+----------+-----------------------------------------------
 1  | cn1.example.com | primary | * running |          | default  | 100      | 16       | host=cn1.example.com user=repmgr dbname=repmgr
 2  | cn2.example.com | standby | ! running |          | default  | 100      | 16       | host=cn2.example.com user=repmgr dbname=repmgr

WARNING: following issues were detected
  - node "cn2.example.com" (ID: 2) is running but the repmgr node record is inactive

From what I can tell from the primary replication stats, replication is progressing normally, although there is some delay, but it is rush hour at the moment.

Both node run:
Debian 10.5
Postgresql 12.4-1
repmgr 5.1.0-1

How can I make cn2 active again? Hopefully without recloning the node since replication seems to continue.
Reply all
Reply to author
Forward
0 new messages